FW: Default MapTransformation...

From: Dodov, Boyko <bdodov_at_nyahnyahspammersnyahnyah>
Date: Wed Jul 20 2011 - 14:57:06 MDT

Hi,

Attached are the source code, the two WRF geo-data domain files (gzipped) and the namelist.wps used to produce the two domain files. The coordinate system is a shifted lat-lon: 39.83329 deg to the North and -98.58331 to the West (and these are assigned respectively to mpCenterLatF and mpCenterLonF). No wrf-specific script has been used for the attached plot, so the default coordinate system must be CE.

I see now why do I plot a uniform grid: mpCenterLatF and mpCenterLonF are the rotation of the coordinate system, while the default parameters (standard parallel and central meridian) are 0, 0. Any idea how to reproduce the rotation of the sphere in ArcGIS?

Thanks,

Boyko

BDY: I am receiving mails from Allison Steiner, Dennis Shea and Gus Correa.

From: David Brown [mailto:dbrown@ucar.edu]
Sent: Wednesday, July 20, 2011 2:39 PM
To: Dodov, Boyko
Subject: Re: Default MapTransformation...

Hi,

Your description of the cylindrical equidistant projection is correct as long as the center latitude (mpCenterLatF in NCL) is set to 0, and the rotation (mpCenterRotF) is also 0. The question is what routine are you using to draw your plot of the grid. Certainly if you just draw a map without any data, you will get a CE plot by default. But if you use one of the provided WRF scripts, supplying the coordinate information from a WRF data file, the script figures out the projection based on metadata in the file. The plot you sent looks to me like a Lambert Conformal projection, one of the standard WRF projections. The fact that the coordinate locations create a rectangle suggests that the projection of your plot is indeed the projection used for the data. But I cannot really say anything more definitive without seeing the actual code you are using.

You are now subscribed to NCL talk, so please send future messages there. Thanks,
 -dave

On Jul 20, 2011, at 11:39 AM, Dodov, Boyko wrote:

Dear Mr. Brown,

I hope you can answer my question about the default map transformation in NCL. I am trying to get the answer of my question directly from you, because signing up for the ncl-talk is taking way too long.

My question is: What is the default map transformation in NCL? The NCL on-line help says:

mpProjection (MapTransformation<http://www.ncl.ucar.edu/Document/HLUs/Classes/MapTransformation.shtml>)
This resource of type NhlTProjection defines the projection used for the map transformation. There are several choices:
Orthographic
Transform using an aziumthal orthographic projection.
Stereographic
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . .
WinkelTripel (Available in version 5.1.0 and later.)
Transform using a WinkelTripel projection.

Default: CylindricalEquidistant

In the Cylindrical Equidistant projection (as described in the textbooks) the cylinder is wrapped around a globe at the equator and ,therefore, all parallels are all strait horizontal lines. In contrast, when using the NCL default projection, this is not the case (see the attached map of my WRF grid made without setting up mpProjection at all). So what is the default projection? Seems like the one used by default is an oblique cylindrical or conical projection.

Thank you in advance for your consideration,

Boyko Dodov

Boyko A. Dodov, Ph. D.
Principal Scientist, Hydrology
AIR Worldwide
131 Dartmouth Street
Boston, MA 02116
+1-617-954-1860
<image001.gif><http://www.air-worldwide.com/>

<grid.jpg>

_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk

Received on Wed Jul 20 14:57:16 2011

This archive was generated by hypermail 2.1.8 : Tue Aug 02 2011 - 14:31:37 MDT